Related Experiment Video
Updated: Jan 5, 2026
A Knowledge Graph Approach to Elucidate the Role of Organellar Pathways in Disease via Biomedical Reports
Published on: October 13, 2023
Inferring Disease-Associated MicroRNAs Using Semi-supervised Multi-Label Graph Convolutional Networks
1Institute of Image Processing and Pattern Recognition, Shanghai Jiao Tong University, and Key Laboratory of System Control and Information Processing, Ministry of Education of China, 200240 Shanghai, China; Department of Medical informatics, Erasmus Medical Center, 3015 CE Rotterdam, the Netherlands.
Abstract:
MicroRNAs (miRNAs) play crucial roles in biological processes involved in diseases. The associations between diseases and protein-coding genes (PCGs) have been well investigated, and miRNAs interact with PCGs to trigger them to be functional. We present a computational method, DimiG, to infer miRNA-associated diseases using a semi-supervised Graph Convolutional Network model (GCN). DimiG uses a multi-label framework to integrate PCG-PCG interactions, PCG-miRNA interactions, PCG-disease associations, and tissue expression profiles. DimiG is trained on disease-PCG associations and an interaction network using a GCN, which is further used to score associations between diseases and miRNAs. We evaluate DimiG on a benchmark set from verified disease-miRNA associations. Our results demonstrate that DimiG outperforms the best unsupervised method and is comparable to two supervised methods. Three case studies of prostate cancer, lung cancer, and inflammatory bowel disease further demonstrate the efficacy of DimiG, where top miRNAs predicted by DimiG are supported by literature.
Insights
We developed DimiG, a computational method using Graph Convolutional Networks, to identify disease-associated microRNAs (miRNAs). DimiG effectively predicts miRNA-disease links by integrating various biological data, aiding disease research.
Area of Science:
- Bioinformatics
- Computational Biology
- Genomics
Background:
- MicroRNAs (miRNAs) are key regulators in biological processes and disease development.
- While protein-coding gene (PCG)-disease associations are known, miRNA-disease links require further investigation.
- miRNAs interact with PCGs, influencing their function and disease pathways.
Purpose of the Study:
- To develop a computational method, DimiG, for inferring miRNA-associated diseases.
- To leverage a semi-supervised Graph Convolutional Network (GCN) model for predicting miRNA-disease associations.
- To integrate diverse biological data for enhanced prediction accuracy.
Main Methods:
- DimiG employs a multi-label GCN framework.
- It integrates PCG-PCG interactions, PCG-miRNA interactions, PCG-disease associations, and tissue expression profiles.
- The model is trained on known disease-PCG associations and interaction networks.
Main Results:
- DimiG effectively scores associations between diseases and miRNAs.
- Performance evaluation on a benchmark dataset shows DimiG outperforms unsupervised methods and rivals supervised methods.
- Case studies on prostate cancer, lung cancer, and inflammatory bowel disease validate DimiG's predictions with existing literature.
Conclusions:
- DimiG is an effective computational tool for predicting miRNA-disease associations.
- The method provides valuable insights into the roles of miRNAs in various diseases.
- DimiG aids in discovering novel miRNA biomarkers and therapeutic targets for diseases.
